今日学术视野()
cond-mat.stat-mech -
cs.AI - 人工智能
cs.CL - 计算与语言
cs.CV - 机器视觉与模式识别
cs.CY - 计算与社会
cs.DC - 分布式、并行与集群计算
cs.IT - 信息论
cs.LG - 自动学习
cs.NA - 数值分析
cs.NE - 神经与进化计算
cs.RO - 机器人学
cs.SI - 社交网络与信息网络
math.OC - 优化与控制
math.ST - 统计理论
physics.soc-ph - 物理学与社会
stat.AP - 应用统计
stat.ME - 统计方法论
stat.ML - (统计)机器学习
& [cond-mat.stat-mech]Universality of 2 1 dimensional RSOS
& [cs.AI]Characterizing Quantifier Fuzzification Mechanisms: a
behavioral guide for practical applications
& [cs.AI]Concept based Attention
& [cs.AI]Learning Bounded Treewidth Bayesian Networks with
Thousands of Variables
& [cs.CL]Machine Comprehension Based on Learning to Rank
& [cs.CL]The Yahoo Query Treebank, V. 1.0
& [cs.CL]Vocabulary Manipulation for Neural Machine
Translation
& [cs.CV]A robust particle detection algorithm based on
& [cs.CV]Action Recognition in Video Using Sparse Coding and
Relative Features
& [cs.CV]Deep Attributes Driven Multi-Camera Person
Re-identification
& [cs.CV]Deep Neural Networks Under Stress
& [cs.CV]Efficiently Creating 3D Training Data for Fine Hand Pose
Estimation
& [cs.CV]Go-ICP: A Globally Optimal Solution to 3D ICP Point-Set
Registration
& [cs.CV]Image-level Classification in Hyperspectral Images using
Feature Descriptors, with Application to Face Recognition
& [cs.CV]On-the-fly Network Pruning for Object Detection
& [cs.CV]Real-time 3D Tracking of Articulated Tools for Robotic
& [cs.CV]Unsupervised Semantic Action Discovery from Video
Collections
& [cs.CV]View Synthesis by Appearance Flow
& [cs.CY]A Novel Framework for Electronic Global Health Record
& [cs.CY]Investigating the opportunities of using mobile learning
by young children in Bulgaria
& [cs.CY]When Do Luxury Cars Hit the Road? Findings by A Big Data
& [cs.DC]Implementation of the open source virtualization
technologies in cloud computing
& [cs.DC]On Storage Allocation for Maximum Service Rate in
Distributed Storage Systems
& [cs.IT]Performance Bounds for Sparse Signal Reconstruction with
Multiple Side Information
& [cs.LG]A constrained L1 minimization approach for estimating
multiple Sparse Gaussian or Nonparanormal Graphical
& [cs.LG]Tweet2Vec: Character-Based Distributed Representations for
Social Media
& [cs.NA]Active Uncertainty Calibration in Bayesian ODE
& [cs.NE]COCO: Performance Assessment
& [cs.RO]A Hierarchical Emotion Regulated Sensorimotor Model: Case
& [cs.RO]Sensorimotor Input as a Language Generalisation Tool: A
Neurorobotics Model for Generation and Generalisation of Noun-Verb
Combinations with Sensorimotor Inputs
& [cs.SI]Multicore-periphery structure in networks
& [cs.SI]Profit-aware Team Grouping in Social Networks: A
Generalized Cover Decomposition Approach
& [math.OC]On the Iteration Complexity of Oblivious First-Order
Optimization Algorithms
& [math.ST]Le cam theory on the comparison of statistical
& [math.ST]On Asymptotics Related to Classical and Bayesian
Inference in Stochastic Differential Equations with Time-Varying
Covariates
& [math.ST]On Classical and Bayesian Asymptotics in Stochastic
Differential Equations with Random Effects having Mixture Normal
Distributions
& [math.ST]On optimality of empirical risk minimization in linear
aggregation
& [math.ST]Power variations and testing for co-jumps: the small
noise approach
& [math.ST]Quantile tests in frequency domain for sinusoid
& [math.ST]The LASSO Estimator: Distributional Properties
& [physics.soc-ph]Race, Religion and the City: Twitter Word
Frequency Patterns Reveal Dominant Demographic Dimensions in the
United States
& [stat.AP]A Poisson process reparameterisation for Bayesian
inference for extremes
& [stat.AP]An Overview of Spatial Econometrics
& [stat.AP]Multi-class Vector AutoRegressive Models for Multi-store
Sales Data
& [stat.ME]Asymptotic equivalence of regularization methods in
thresholded parameter space
& [stat.ME]Asymptotic properties for combined $L_1$ and concave
regularization
& [stat.ME]High dimensional thresholded regression and shrinkage
& [stat.ME]Innovated scalable efficient estimation in ultra-large
Gaussian graphical models
& [stat.ME]Interaction pursuit in high-dimensional multi-response
regression via distance correlation
& [stat.ME]Nonparametric hierarchical Bayesian quantiles
& [stat.ME]The Distance Precision Matrix: computing networks from
nonlinear relationships
& [stat.ME]The constrained Dantzig selector with enhanced
consistency
& [stat.ME]Tuning parameter selection in high dimensional penalized
likelihood
& [stat.ME]What’s the value in more moments?
& [stat.ML]Generalized Sparse Precision Matrix Selection for
Fitting Multivariate Gaussian Random Fields to Large Data
& [stat.ML]Random forests for survival analysis using maximally
selected rank statistics&
·····································
& [cond-mat.stat-mech]Universality of 2 1
dimensional RSOS models
Jeffrey Kelling,
G&za &Odor, Sibylle Gemming
http://arxiv.org/abs/v1&
Extensive dynamical
simulations of Restricted Solid on Solid models in $D=2 1$
dimensions have been done using parallel multisurface algorithms
implemented on graphics cards. Numerical evidence is presented that
these models exhibit KPZ surface growth scaling, irrespective of
the step heights $N$. We show that by increasing $N$ the
corrections to scaling increase, thus smaller step sized models
describe better the asymptotic, long wave scaling
behavior.&
& [cs.AI]Characterizing Quantifier
Fuzzification Mechanisms: a behavioral guide for practical
applications
F. Diaz-Hermida, M.
Pereira-Fari&a, Juan C. Vidal, A. Ramos-Soto
http://arxiv.org/abs/v1&
Important advances
have been made in the fuzzy quantification field. Nevertheless,
some problems remain when we face the decision of selecting the
most convenient model for a specific application. In the
literature, several desirable adequacy properties have been
proposed, but theoretical limits impede quantification models from
simultaneously fulfilling every adequacy property that has been
defined. Besides, the complexity of model definitions and adequacy
properties makes very difficult for real users to understand the
particularities of the different models that have been presented.
In this work we will present several criteria conceived to help in
the process of selecting the most adequate Quantifier Fuzzification
Mechanisms for specific practical applications. In addition, some
of the best known well-behaved models will be compared against this
list of criteria. Based on this analysis, some guidance to choose
fuzzy quantification models for practical applications will be
provided.&
& [cs.AI]Concept
based Attention
Jie You, Xin Yang,
Matthias Hub
http://arxiv.org/abs/v1&
Attention endows
animals an ability to concentrate on the most relevant information
among a deluge of distractors at any given time, either through
volitionally ‘top-down’ biasing, or driven by automatically
‘bottom-up’ saliency of stimuli, in favour of advantageous
competition in neural modulations for information processing.
Nevertheless, instead of being limited to perceive simple features,
human and other advanced animals adaptively learn the world into
categories and abstract concepts from experiences, imparting the
world meanings. This thesis suggests that the high-level cognitive
ability of human is more likely driven by attention basing on
abstract perceptions, which is defined as concept based attention
& [cs.AI]Learning Bounded Treewidth
Bayesian Networks with Thousands of Variables
Mauro Scanagatta,
Giorgio Corani, Cassio P. de Campos, Marco Zaffalon
http://arxiv.org/abs/v1&
We present a method
for learning treewidth-bounded Bayesian networks from data sets
containing thousands of variables. Bounding the treewidth of a
Bayesian greatly reduces the complexity of inferences. Yet, being a
global property of the graph, it considerably increases the
difficulty of the learning process. We propose a novel algorithm
for this task, able to scale to large domains and large treewidths.
Our novel approach consistently outperforms the state of the art on
data sets with up to ten thousand variables.&
& [cs.CL]Machine
Comprehension Based on Learning to Rank
Tian Tian, Yuezhang
http://arxiv.org/abs/v1&
Machine comprehension
plays an essential role in NLP and has been widely explored with
dataset like MCTest. However, this dataset is too simple and too
small for learning true reasoning abilities.
\cite{hermann2015teaching} therefore release a large scale news
article dataset and propose a deep LSTM reader system for machine
comprehension. However, the training process is expensive. We
therefore try feature-engineered approach with semantics on the new
dataset to see how traditional machine learning technique and
semantics can help with machine comprehension. Meanwhile, our
proposed L2R reader system achieves good performance with
efficiency and less training data.&
& [cs.CL]The
Yahoo Query Treebank, V. 1.0
Yuval Pinter, Roi
Reichart, Idan Szpektor
http://arxiv.org/abs/v2&
A description and
annotation guidelines for the Yahoo Webscope release of Query
Treebank, Version 1.0, May 2016.&
& [cs.CL]Vocabulary Manipulation for
Neural Machine Translation
Haitao Mi, Zhiguo
Wang, Abe Ittycheriah
http://arxiv.org/abs/v1&
In order to capture
rich language phenomena, neural machine translation models have to
use a large vocabulary size, which requires high computing time and
large memory usage. In this paper, we alleviate this issue by
introducing a sentence-level or batch-level vocabulary, which is
only a very small sub-set of the full output vocabulary. For each
sentence or batch, we only predict the target words in its
sentence-level or batch-level vocabulary. Thus, we reduce both the
computing time and the memory usage. Our method simply takes into
account the translation options of each word or phrase in the
source sentence, and picks a very small target vocabulary for each
sentence based on a word-to-word translation model or a bilingual
phrase library learned from a traditional machine translation
model. Experimental results on the large-scale English-to-French
task show that our method achieves better translation performance
by 1 BLEU point over the large vocabulary neural machine
translation system of Jean et al. (2015).&
& [cs.CV]A
robust particle detection algorithm based on
Alvaro Rodriguez,
Hanqing Zhang, Krister Wiklund, Tomas Brodin, Jonatan Klaminder,
Patrik Andersson, Magnus Andersson
http://arxiv.org/abs/v1&
Particle tracking is
common in many biophysical, ecological, and micro-fluidic
applications. Reliable tracking information is heavily dependent on
of the system under study and algorithms that correctly determines
particle position between images. However, in a real environmental
context with the presence of noise including particular or
dissolved matter in water, and low and fluctuating light
conditions, many algorithms fail to obtain reliable information. We
propose a new algorithm, the Circular Symmetry algorithm (C-Sym),
for detecting the position of a circular particle with high
accuracy and precision in noisy conditions. The algorithm takes
advantage of the spatial symmetry of the particle allowing for
subpixel accuracy. We compare the proposed algorithm with four
different methods using both synthetic and experimental datasets.
The results show that C-Sym is the most accurate and precise
algorithm when tracking micro-particles in all tested conditions
and it has the potential for use in applications including tracking
biota in their environment.&
& [cs.CV]Action
Recognition in Video Using Sparse Coding and Relative
Anali Alfaro,
Domingo Mery, Alvaro Soto
http://arxiv.org/abs/v1&
This work presents an
approach to category-based action recognition in video using sparse
coding techniques. The proposed approach includes two main
contributions: i) A new method to handle intra-class variations by
decomposing each video into a reduced set of representative atomic
action acts or key-sequences, and ii) A new video descriptor, ITRA:
Inter-Temporal Relational Act Descriptor, that exploits the power
of comparative reasoning to capture relative similarity relations
among key-sequences. In terms of the method to obtain
key-sequences, we introduce a loss function that, for each video,
leads to the identification of a sparse set of representative
key-frames capturing both, relevant particularities arising in the
input video, as well as relevant generalities arising in the
complete class collection. In terms of the method to obtain the
ITRA descriptor, we introduce a novel scheme to quantify relative
intra and inter-class similarities among local temporal patterns
arising in the videos. The resulting ITRA descriptor demonstrates
to be highly effective to discriminate among action categories. As
a result, the proposed approach reaches remarkable action
recognition performance on several popular benchmark datasets,
outperforming alternative state-of-the-art techniques by a large
& [cs.CV]Deep
Attributes Driven Multi-Camera Person
Re-identification
Chi Su, Shiliang
Zhang, Junliang Xing, Wen Gao, Qi Tian
http://arxiv.org/abs/v1&
The visual appearance
of a person is easily affected by many factors like pose
variations, viewpoint changes and camera parameter differences.
This makes person Re-Identification (ReID) among multiple cameras a
very challenging task. This work is motivated to learn mid-level
human attributes which are robust to such visual appearance
variations. And we propose a semi-supervised attribute learning
framework which progressively boosts the accuracy of attributes
only using a limited number of labeled data. Specifically, this
framework involves a three-stage training. A deep Convolutional
Neural Network (dCNN) is first trained on an independent dataset
labeled with attributes. Then it is fine-tuned on another dataset
only labeled with person IDs using our defined triplet loss.
Finally, the updated dCNN predicts attribute labels for the target
dataset, which is combined with the independent dataset for the
final round of fine-tuning. The predicted attributes, namely
\emph{deep attributes} exhibit superior generalization ability
across different datasets. By directly using the deep attributes
with simple Cosine distance, we have obtained surprisingly good
accuracy on four person ReID datasets. Experiments also show that a
simple metric learning modular further boosts our method, making it
significantly outperform many recent works.&
& [cs.CV]Deep
Neural Networks Under Stress
Micael Carvalho,
Matthieu Cord, Sandra Avila, Nicolas Thome, Eduardo
http://arxiv.org/abs/v1&
In recent years, deep
architectures have been used for transfer learning with
state-of-the-art performance in many datasets. The properties of
their features remain, however, largely unstudied under the
transfer perspective. In this work, we present an extensive
analysis of the resiliency of feature vectors extracted from deep
models, with special focus on the trade-off between performance and
compression rate. By introducing perturbations to image
descriptions extracted from a deep convolutional neural network, we
change their precision and number of dimensions, measuring how it
affects the final score. We show that deep features are more robust
to these disturbances when compared to classical approaches,
achieving a compression rate of 98.4%, while losing only 0.88% of
their original score for Pascal VOC 2007.&
& [cs.CV]Efficiently Creating 3D
Training Data for Fine Hand Pose Estimation
Markus Oberweger,
Gernot Riegler, Paul Wohlhart, Vincent Lepetit
http://arxiv.org/abs/v1&
While many recent
hand pose estimation methods critically rely on a training set of
labelled frames, the creation of such a dataset is a challenging
task that has been overlooked so far. As a result, existing
datasets are limited to a few sequences and individuals, with
limited accuracy, and this prevents these methods from delivering
their full potential. We propose a semi-automated method for
efficiently and accurately labeling each frame of a hand depth
video with the corresponding 3D locations of the joints: The user
is asked to provide only an estimate of the 2D reprojections of the
visible joints in some reference frames, which are automatically
selected to minimize the labeling work by efficiently optimizing a
sub-modular loss function. We then exploit spatial, temporal, and
appearance constraints to retrieve the full 3D poses of the hand
over the complete sequence. We show that this data can be used to
train a recent state-of-the-art hand pose estimation method,
leading to increased accuracy. The code and dataset can be found on
our website
https://cvarlab.icg.tugraz.at/projects/hand_detection/&
& [cs.CV]Go-ICP:
A Globally Optimal Solution to 3D ICP Point-Set
Registration
Jiaolong Yang,
Hongdong Li, Dylan Campbell, Yunde Jia
http://arxiv.org/abs/v1&
The Iterative Closest
Point (ICP) algorithm is one of the most widely used methods for
point-set registration. However, being based on local iterative
optimization, ICP is known to be susceptible to local minima. Its
performance critically relies on the quality of the initialization
and only local optimality is guaranteed. This paper presents the
first globally optimal algorithm, named Go-ICP, for Euclidean
(rigid) registration of two 3D point-sets under the L2 error metric
defined in ICP. The Go-ICP method is based on a branch-and-bound
(BnB) scheme that searches the entire 3D motion space SE(3). By
exploiting the special structure of SE(3) geometry, we derive novel
upper and lower bounds for the registration error function. Local
ICP is integrated into the BnB scheme, which speeds up the new
method while guaranteeing global optimality. We also discuss
extensions, addressing the issue of outlier robustness. The
evaluation demonstrates that the proposed method is able to produce
reliable registration results regardless of the initialization.
Go-ICP can be applied in scenarios where an optimal solution is
desirable or where a good initialization is not always
available.&
& [cs.CV]Image-level Classification in
Hyperspectral Images using Feature Descriptors, with Application to
Face Recognition
Vivek Sharma, Luc
http://arxiv.org/abs/v1&
In this paper, we
proposed a novel pipeline for image-level classification in the
hyperspectral images. By doing this, we show that the
discriminative spectral information at image-level features lead to
significantly improved performance in a face recognition task. We
also explored the potential of traditional feature descriptors in
the hyperspectral images. From our evaluations, we observe that
SIFT features outperform the state-of-the-art hyperspectral face
recognition methods, and also the other descriptors. With the
increasing deployment of hyperspectral sensors in a multitude of
applications, we believe that our approach can effectively exploit
the spectral information in hyperspectral images, thus beneficial
to more accurate classification.&
& [cs.CV]On-the-fly Network Pruning for
Object Detection
Marc Masana, Joost
van de Weijer, Andrew D. Bagdanov
http://arxiv.org/abs/v1&
Object detection with
deep neural networks is often performed by passing a few thousand
candidate bounding boxes through a deep neural network for each
image. These bounding boxes are highly correlated since they
originate from the same image. In this paper we investigate how to
exploit feature occurrence at the image scale to prune the neural
network which is subsequently applied to all bounding boxes. We
show that removing units which have near-zero activation in the
image allows us to significantly reduce the number of parameters in
the network. Results on the PASCAL 2007 Object Detection Challenge
demonstrate that up to 40% of units in some fully-connected layers
can be entirely eliminated with little change in the detection
& [cs.CV]Real-time 3D Tracking of
Articulated Tools for Robotic Surgery
Menglong Ye, Lin
Zhang, Stamatia Giannarou, Guang-Zhong Yang
http://arxiv.org/abs/v1&
In robotic surgery,
tool tracking is important for providing safe tool-tissue
interaction and facilitating surgical skills assessment. Despite
recent advances in tool tracking, existing approaches are faced
with major difficulties in real-time tracking of articulated tools.
Most algorithms are tailored for offline processing with
pre-recorded videos. In this paper, we propose a real-time 3D
tracking method for articulated tools in robotic surgery. The
proposed method is based on the CAD model of the tools as well as
robot kinematics to generate online part-based templates for
efficient 2D matching and 3D pose estimation. A robust verification
approach is incorporated to reject outliers in 2D detections, which
is then followed by fusing inliers with robot kinematic readings
for 3D pose estimation of the tool. The proposed method has been
validated with phantom data, as well as ex vivo and in vivo
experiments. The results derived clearly demonstrate the
performance advantage of the proposed method when compared to the
state-of-the-art.&
& [cs.CV]Unsupervised Semantic Action
Discovery from Video Collections
Ozan Sener, Amir
Roshan Zamir, Chenxia Wu, Silvio Savarese, Ashutosh
http://arxiv.org/abs/v1&
Human communication
takes many forms, including speech, text and instructional videos.
It typically has an underlying structure, with a starting point,
ending, and certain objective steps between them. In this paper, we
consider instructional videos where there are tens of millions of
them on the Internet. We propose a method for parsing a video into
such semantic steps in an unsupervised way. Our method is capable
of providing a semantic “storyline” of the video composed of its
objective steps. We accomplish this using both visual and language
cues in a joint generative model. Our method can also provide a
textual description for each of the identified semantic steps and
video segments. We evaluate our method on a large number of complex
YouTube videos and show that our method discovers semantically
correct instructions for a variety of tasks.&
& [cs.CV]View
Synthesis by Appearance Flow
Tinghui Zhou,
Shubham Tulsiani, Weilun Sun, Jitendra Malik, Alexei A.
http://arxiv.org/abs/v1&
Given one or more
images of an object (or a scene), is it possible to synthesize a
new image of the same instance observed from an arbitrary
viewpoint? In this paper, we attempt to tackle this problem, known
as novel view synthesis, by re-formulating it as a pixel copying
task that avoids the notorious difficulties of generating pixels
from scratch. Our approach is built on the observation that the
visual appearance of different views of the same instance is highly
correlated. Such correlation could be explicitly learned by
training a convolutional neural network (CNN) to predict appearance
flows & 2-D coordinate vectors specifying which pixels in the input
view could be used to reconstruct the target view. We show that for
both objects and scenes, our approach is able to generate
higher-quality synthesized views with crisp texture and boundaries
than previous CNN-based techniques.&
& [cs.CY]A Novel
Framework for Electronic Global Health Record
Nael A. H AbuOun,
Ayman Abdel-Hamid, Mohamad Abou El-Nasr
http://arxiv.org/abs/v1&
When most patients
visit physicians in a clinic or a hospital, they are asked about
their medical history and related medical tests' results which
might not exist or might simply have been lost over time. In
emergency situations, many patients suffer or sadly die because of
lack of pertinent medical information. Patient’s Health information
(PHI) saved by Electronic Medical Record (EMR) could be accessible
only by a hospital using their EMR system. Furthermore, Personal
Health Record (PHR) information cannot be solely relied on since it
is controlled solely by patients. This paper introduces a novel
framework for accessing, sharing, and controlling the medical
records for patients and their physicians globally, while patients'
PHI are securely stored and their privacy is taken into
consideration. Based on the framework, a proof of concept prototype
is implemented. Preliminary performance evaluation results indicate
the validity and viability of the proposed
framework.&
& [cs.CY]Investigating the
opportunities of using mobile learning by young children in
Radoslava Kraleva,
Aleksandar Stoimenovski, Dafina Kostadinova, Velin
http://arxiv.org/abs/v1&
This paper provides
an analysis of literature related to the use of mobile devices in
teaching young children. For this purpose, the most popular mobile
operating systems in Bulgaria are considered and the functionality
of the existing mobile applications with Bulgarian interface is
discussed. The results of a survey of parents' views regarding the
mobile devices as a learning tool are presented and the ensuing
conclusions are provided.&
& [cs.CY]When Do
Luxury Cars Hit the Road? Findings by A Big Data
Yang Feng, Jiebo
http://arxiv.org/abs/v2&
In this paper, we
focus on studying the appearing time of different kinds of cars on
the road. This information will enable us to infer the life style
of the car owners. The results can further be used to guide
marketing towards car owners. Conventionally, this kind of study is
carried out by sending out questionnaires, which is limited in
scale and diversity. To solve this problem, we propose a fully
automatic method to carry out this study. Our study is based on
publicly available surveillance camera data. To make the results
reliable, we only use the high resolution cameras (i.e. resolution
greater than ). Images from
the public cameras are downloaded every minute. After obtaining
50,000 images, we apply faster R-CNN (region-based convoluntional
neural network) to detect the cars in the downloaded images and a
fine-tuned VGG16 model is used to recognize the car makes. Based on
the recognition results, we present a data-driven analysis on the
relationship between car makes and their appearing times, with
implications on lifestyles.&
& [cs.DC]Implementation of the open
source virtualization technologies in cloud
Mohammad Mamun Or
Rashid, M. Masud Rana, Jugal Krishna Das
http://arxiv.org/abs/v1&
The Virtualization
and Cloud Computing is a recent buzzword in the digital world.
Cloud computing provide IT as a service to the users on demand
basis. This service has greater flexibility, availability,
reliability and scalability with utility computing model. This new
concept of computing has an immense potential in it to be used in
the field of e-governance and in the overall IT development
perspective in developing countries like
Bangladesh.&
& [cs.DC]On
Storage Allocation for Maximum Service Rate in Distributed Storage
Moslem Noori, Emina
Soljanin, Masoud Ardakani
http://arxiv.org/abs/v1&
Storage allocation
affects important performance measures of distributed storage
systems. Most previous studies on the storage allocation consider
its effect separately either on the success of the data recovery or
on the service rate (time) where it is assumed that no access
failure happens in the system. In this paper, we go one step
further and incorporate the access model and the success of data
recovery into the service rate analysis. In particular, we focus on
quasi-uniform storage allocation and provide a service rate
analysis for both fixed-size and probabilistic access models at the
nodes. Using this analysis, we then show that for the case of
exponential waiting time distribution at individuals storage nodes,
minimal spreading allocation results in the highest system service
rate for both access models. This means that for a given storage
budget, replication provides a better service rate than a coded
storage solution.&
& [cs.IT]Performance Bounds for Sparse
Signal Reconstruction with Multiple Side
Information
Huynh Van Luong,
Jurgen Seiler, Andre Kaup, Soren Forchhammer, Nikos
Deligiannis
http://arxiv.org/abs/v1&
In the context of
compressive sensing (CS), this paper considers the problem of
reconstructing sparse signals with the aid of other given
correlated sources as multiple side information (SI). To address
this problem, we propose a reconstruction algorithm with multiple
SI (RAMSI) that solves a general weighted $n$-$\ell{1}$ norm minimization. The
proposed RAMSI algorithm takes advantage of both CS and the
$n$-$\ell{1}$ minimization by adaptively computing optimal
weights among SI signals at every reconstructed iteration. In
addition, we establish theoretical performance bounds on the number
of measurements that are required to successfully reconstruct the
original sparse source using RAMSI under arbitrary support SI
conditions. The analyses of the established bounds reveal that
RAMSI can achieve sharper bounds and significant performance
improvements compared to classical CS. We evaluate experimentally
the proposed algorithm and the established bounds using synthetic
sparse signals as well as correlated feature histograms, extracted
from a multiview image database for object recognition. The
obtained results show clearly that the proposed RAMSI algorithm
outperforms classical CS and CS with single SI in terms of both the
theoretical bounds and the practical
performance.&
& [cs.LG]A
constrained L1 minimization approach for estimating multiple Sparse
Gaussian or Nonparanormal Graphical Models
Beilun Wang,
Ritambhara Singh, Yanjun Qi
http://arxiv.org/abs/v1&
The flood of
multi-context measurement data from many scientific domains have
created an urgent need to reconstruct context-specific variable
networks, that can significantly simplify network-driven studies.
Computationally, this problem can be formulated as jointly
estimating multiple different, but related, sparse Undirected
Graphical Models (UGM) from samples aggregated across several
tasks. Previous joint-UGM studies could not fully address the
challenge since they mostly focus on Gaussian Graphical Models
(GGM) and have used likelihood-based formulations to push multiple
estimated networks toward a common pattern. Differently, we propose
a novel approach, SIMULE (learning Shared and Individual parts of
MULtiple graphs Explicitly) to solve multi-task UGM using a l1
constrained optimization. SIMULE can handle both multivariate
Gaussian and multivariate Nonparanormal data (greatly relaxing the
normality assumption most real data do not follow). SIMULE is cast
as independent subproblems of linear programming that can be solved
efficiently. It automatically infers specific dependencies that are
unique to each context as well as shared substructures preserved
among all the contexts. Theoretically we prove that SIMULE achieves
a consistent estimation at rate O(log(Kp)/ntot) (not been proved
before). On four synthetic datasets and two real datasets, SIMULE
shows significant improvements over state-of-the-art multi-sGGM and
single-UGM baselines.&
& [cs.LG]Tweet2Vec: Character-Based
Distributed Representations for Social Media
Bhuwan Dhingra,
Zhong Zhou, Dylan Fitzpatrick, Michael Muehl, William W.
http://arxiv.org/abs/v1&
Text from social
media provides a set of challenges that can cause traditional NLP
approaches to fail. Informal language, spelling errors,
abbreviations, and special characters are all commonplace in these
posts, leading to a prohibitively large vocabulary size for
word-level approaches. We propose a character composition model,
tweet2vec, which finds vector-space representations of whole tweets
by learning complex, non-local dependencies in character sequences.
The proposed model outperforms a word-level baseline at predicting
user-annotated hashtags associated with the posts, doing
significantly better when the input contains many out-of-vocabulary
words or unusual character sequences. Our tweet2vec encoder is
publicly available.&
& [cs.NA]Active
Uncertainty Calibration in Bayesian ODE Solvers
Hans Kersting,
Philipp Hennig
http://arxiv.org/abs/v1&
There is resurging
interest, in statistics and machine learning, in solvers for
ordinary differential equations (ODEs) that return probability
measures instead of point estimates. Recently, Conrad et al.
introduced a sampling-based class of methods that are
‘well-calibrated’ in a specific sense. But the computational cost
of these methods is significantly above that of classic methods. On
the other hand, Schober et al. pointed out a precise connection
between classic Runge-Kutta ODE solvers and Gaussian filters, which
gives only a rough probabilistic calibration, but at negligible
cost overhead. By formulating the solution of ODEs as approximate
inference in linear Gaussian SDEs, we investigate a range of
probabilistic ODE solvers, that bridge the trade-off between
computational cost and probabilistic calibration, and identify the
inaccurate gradient measurement as the crucial source of
uncertainty. We propose the novel filtering-based method Bayesian
Quadrature filtering (BQF) which uses Bayesian quadrature to
actively learn the imprecision in the gradient measurement by
collecting multiple gradient evaluations.&
& [cs.NE]COCO:
Performance Assessment
Nikolaus Hansen,
Anne Auger, Dimo Brockhoff, Dejan Tu&ar, Tea Tu&ar
http://arxiv.org/abs/v1&
We present an
any-time performance assessment for benchmarking numerical
optimization algorithms in a black-box scenario, applied within the
COCO benchmarking platform. The performance assessment is based on
runtimes measured in number of objective function evaluations to
reach one or several quality indicator target values. We argue that
runtime is the only available measure with a generic, meaningful,
and quantitative interpretation. We discuss the choice of the
target values, runlength-based targets, and the aggregation of
results by using simulated restarts, averages, and empirical
distribution functions.&
& [cs.RO]A
Hierarchical Emotion Regulated Sensorimotor Model: Case
Junpei Zhong, Rony
Novianto, Mingjun Dai, Xinzheng Zhang, Angelo
http://arxiv.org/abs/v1&
Inspired by the
hierarchical cognitive architecture and the perception-action model
(PAM), we propose that the internal status acts as a kind of
common-coding representation which affects, mediates and even
regulates the sensorimotor behaviours. These regulation can be
depicted in the Bayesian framework, that is why cognitive agents
are able to generate behaviours with subtle differences according
to their emotion or recognize the emotion by perception. A novel
recurrent neural network called recurrent neural network with
parametric bias units (RNNPB) runs in three modes, constructing a
two-level emotion regulated learning model, was further applied to
testify this theory in two different cases.&
& [cs.RO]Sensorimotor Input as a
Language Generalisation Tool: A Neurorobotics Model for Generation
and Generalisation of Noun-Verb Combinations with Sensorimotor
Junpei Zhong, Martin
Peniak, Jun Tani, Tetsuya Ogata, Angelo Cangelosi
http://arxiv.org/abs/v1&
The paper presents a
neurorobotics cognitive model to explain the understanding and
generalisation of nouns and verbs combinations when a vocal command
consisting of a verb-noun sentence is provided to a humanoid robot.
This generalisation process is done via the grounding process:
different objects are being interacted, and associated, with
different motor behaviours, following a learning approach inspired
by developmental language acquisition in infants. This cognitive
model is based on Multiple Time-scale Recurrent Neural Networks
(MTRNN).With the data obtained from object manipulation tasks with
a humanoid robot platform, the robotic agent implemented with this
model can ground the primitive embodied structure of verbs through
training with verb-noun combination samples. Moreover, we show that
a functional hierarchical architecture, based on MTRNN, is able to
generalise and produce novel combinations of noun-verb sentences.
Further analyses of the learned network dynamics and
representations also demonstrate how the generalisation is possible
via the exploitation of this functional hierarchical recurrent
& [cs.SI]Multicore-periphery structure
in networks
Bowen Yan, Jianxi
http://arxiv.org/abs/v1&
Many real-world
networked systems exhibit a multicore-periphery structure, i.e.,
multiple cores, each of which contains densely connected elements,
surrounded by sparsely connected elements that define the
periphery. Identification of the multiple-periphery structure can
provide a new handle on structures and functions of various complex
networks, such as cognitive and biological networks, food webs,
social networks, and communication and transportation networks.
However, still no quantitative method exists to identify the
multicore-periphery structure embedded in networks. Prior studies
on core-periphery structure focused on either dichotomous or
continuous division of a network into a single core and a
periphery, whereas community detection algorithms did not discern
the periphery from dense cohesive communities. Herein, we introduce
a method to identify the optimal partition of a network into
multiple dense cores and a loosely-connected periphery, and test
the method on a well-known social network and the technology space
network, which are best characterized by multiple-core structures.
Our method gives precise and meaningful results. The analysis of
multicore-periphery structure may advance our understandings of the
structures and functions in diverse real-world
networks.&
& [cs.SI]Profit-aware Team Grouping in
Social Networks: A Generalized Cover Decomposition
http://arxiv.org/abs/v1&
In this paper, we
investigate the profit-aware team grouping problem in social
networks. We consider a setting in which people possess different
skills and compatibility among these individuals is captured by a
social network. Here, we assume a collection of tasks, where each
task requires a specific set of skills, and yields a different
profit upon completion. Active and qualified individuals may
collaborate with each other in the form of \emph{teams} to
accomplish a set of tasks. Our goal is to find a grouping method
that maximizes the total profit of the tasks that these teams can
complete. Any feasible grouping must satisfy the following three
conditions: (i) each team possesses all skills required by the
task, (ii) individuals within the same team are social compatible,
and (iii) each individual is not overloaded. We refer to this as
the \textsc{TeamGrouping} problem. Our work presents a detailed
analysis of the computational complexity of the problem, and
propose a LP-based approximation algorithm to tackle it and its
variants. Although we focus on team grouping in this paper, our
results apply to a broad range of optimization problems that can be
formulated as a cover decomposition problem.&
& [math.OC]On
the Iteration Complexity of Oblivious First-Order Optimization
Algorithms
Yossi Arjevani, Ohad
http://arxiv.org/abs/v1&
We consider a broad
class of first-order optimization algorithms which are
\emph{oblivious}, in the sense that their step sizes are scheduled
regardless of the function under consideration, except for limited
side-information such as smoothness or strong convexity parameters.
With the knowledge of these two parameters, we show that any such
algorithm attains an iteration complexity lower bound of
$\Omega(\sqrt{L/\epsilon})$ for $L$-smooth convex functions, and
$\tilde{\Omega}(\sqrt{L/\mu}\ln(1/\epsilon))$ for $L$-smooth
$\mu$-strongly convex functions. These lower bounds are stronger
than those in the traditional oracle model, as they hold
independently of the dimension. To attain these, we abandon the
oracle model in favor of a structure-based approach which builds
upon a framework recently proposed in (Arjevani et al., 2015). We
further show that without knowing the strong convexity parameter,
it is impossible to attain an iteration complexity better than
$\tilde{\Omega}\left((L/\mu)\ln(1/\epsilon)\right)$. This result is
then used to formalize an observation regarding $L$-smooth convex
functions, namely, that the iteration complexity of algorithms
employing time-invariant step sizes must be at least
$\Omega(L/\epsilon)$.&
& [math.ST]Le
cam theory on the comparison of statistical
http://arxiv.org/abs/v1&
We recall the main
concepts of the Le Cam theory of statistical experiments ,
especially the notion of Le Cam distance and its properties. We
also review classical tools for bounding such a distance before
presenting some examples. A proof of the classical equivalence
result between density estimation problems and Gaussian white noise
models will be analyzed.&
& [math.ST]On
Asymptotics Related to Classical and Bayesian Inference in
Stochastic Differential Equations with Time-Varying
Covariates
Trisha Maitra,
Sourabh Bhattacharya
http://arxiv.org/abs/v1&
Delattre et al.
(2013) considered n independent stochastic differential equations
(SDEs), where in each case the drift term is modeled by a random
effect times a known function free of parameters. The distribution
of the random effects are assumed to depend upon unknown parameters
which are to be learned about. Assuming the independent and
identical (iid) situation the authors provide independent proofs of
consistency and asymptotic normality of the maximum likelihood
estimators (MLEs) of the hyper-parameters of their random effects
parameters. In this article, we generalize the random effect term
by incorporating time-dependent covariates and consider both fixed
and random effects set-ups. We also allow the functional part
associated with the drift function to depend upon unknown
parameters. In this general set-up of SDE system we establish
consistency and asymptotic normality of the MLE through
verification of the regularity conditions required by existing
relevant theorems. Besides, we consider the Bayesian approach to
learning about the population parameters, and prove consistency and
asymptotic normality of the corresponding posterior
distribution.&
& [math.ST]On
Classical and Bayesian Asymptotics in Stochastic Differential
Equations with Random Effects having Mixture Normal
Distributions
Trisha Maitra,
Sourabh Bhattacharya
http://arxiv.org/abs/v1&
Delattre et al.
(2013) considered a system of stochastic differential equations
(SDEs) in a random effects set-up. Under the independent and
identical (iid) situation, and assuming normal distribution of the
random effects, they established weak consistency and asymptotic
normality of the maximum likelihood estimators (MLEs) of the
population parameters of the random effects. In this article,
respecting the increasing importance and versatility of normal
mixtures and their ability to approximate any standard
distribution, we consider the random effects having finite mixture
of normal distributions and prove asymptotic results associated
with the MLEs in both independent and identical (iid) and
independent but not identical (non-iid) situations. Besides, we
consider iid and non-iid set-ups under the Bayesian paradigm and
establish posterior consistency and asymptotic normality of the
posterior distribution of the population
parameters.&
& [math.ST]On
optimality of empirical risk minimization in linear
aggregation
http://arxiv.org/abs/v1&
In the first part of
this paper, we show that the small-ball condition, recently
introduced by Mendelson (2015), may behave poorly for important
classes of localized functions such as wavelets, leading to
suboptimal estimates of the rate of convergence of ERM for the
linear aggregation problem. In a second part, we derive optimal
upper and lower bounds for the excess risk of ERM when the
dictionary is made of trigonometric functions. While the validity
of the small-ball condition remains essentially open in the Fourier
case, we show strong connection between our results and
concentration inequalities recently obtained for the excess risk in
Chatterjee (2014) and van de Geer and Wainwright
& [math.ST]Power
variations and testing for co-jumps: the small noise
http://arxiv.org/abs/v2&
In this paper we
study the effects of noise on the bipower variation (BPV), realized
volatility (RV) and testing for co-jumps in high-frequency data
under the small noise framework. We first establish asymptotic
properties of the BPV in this framework. In the presence of the
small noise, the RV is asymptotically biased and the additional
asymptotic conditional variance term appears in its limit
distribution. We also give feasible estimation methods of the
asymptotic conditional variances of the RV. Second, we derive the
asymptotic distribution of the test statistic proposed in Jacod and
Todorov(2009) under the presence of small noise for testing the
presence of co-jumps in two dimensional It^o semimartingale. In
contrast to the setting in Jacod and Todorov(2009), we show that
the additional conditional asymptotic variance terms appear, and
give consistent estimation procedures for the asymptotic
conditional variances in order to make the test feasible.
Simulation experiments show that our asymptotic results give
reasonable approximations in the finite sample
& [math.ST]Quantile tests in frequency
domain for sinusoid models
http://arxiv.org/abs/v1&
For second order
stationary processes, the spectral distribution function is
uniquely deter- mined by the autocovariance functions of the
processes. We define the quantiles of the spectral distribution
function and propose two estimators for the quantiles. Asymptotic
properties of both estimators are elucidated and the difference
from the quantile estimators in time do- main is also indicated. We
construct a testing procedure of quantile tests from the asymptotic
distribution of the estimators and strong statistical power is
shown in our numerical studies.&
& [math.ST]The
LASSO Estimator: Distributional Properties
Rakshith Jagannath,
Neelesh S Upadhye
http://arxiv.org/abs/v1&
The least absolute
shrinkage and selection operator (LASSO) is a popular technique for
simultaneous estimation and model selection. There have been a lot
of studies on the large sample asymptotic distributional properties
of the LASSO estimator, but it is also well-known that the
asymptotic results can give a wrong picture of the LASSO
estimator’s actual finite-sample behavior. The finite sample
distribution of the LASSO estimator has been previously studied for
the special case of orthogonal models. The aim in this work is to
generalize the finite sample distribution properties of LASSO
estimator for a real and linear measurement model in Gaussian
noise. In this work, we derive an expression for the finite sample
characteristic function of the LASSO estimator, we then use the
Fourier slice theorem to obtain an approximate expression for the
marginal probability density functions of the one-dimensional
components of a linear transformation of the LASSO
estimator.&
& [physics.soc-ph]Race, Religion and the City:
Twitter Word Frequency Patterns Reveal Dominant Demographic
Dimensions in the United States
Eszter Bok&nyi,
D&niel Kondor, L&szl& Dobos, Tam&s Sebők, J&zsef St&ger, Istv&n
Csabai, G&bor Vattay
http://arxiv.org/abs/v2&
Recently, numerous
approaches have emerged in the social sciences to exploit the
opportunities made possible by the vast amounts of data generated
by online social networks (OSNs). Having access to information
about users on such a scale opens up a range of possibilities, all
without the limitations associated with often slow and expensive
paper-based polls. A question that remains to be satisfactorily
addressed, however, is how demography is represented in the OSN
content? Here, we study language use in the US using a corpus of
text compiled from over half a billion geo-tagged messages from the
online microblogging platform Twitter. Our intention is to reveal
the most important spatial patterns in language use in an
unsupervised manner and relate them to demographics. Our approach
is based on Latent Semantic Analysis (LSA) augmented with the
Robust Principal Component Analysis (RPCA) methodology. We find
spatially correlated patterns that can be interpreted based on the
words associated with them. The main language features can be
related to slang use, urbanization, travel, religion and ethnicity,
the patterns of which are shown to correlate plausibly with
traditional census data. Our findings thus validate the concept of
demography being represented in OSN language use and show that the
traits observed are inherently present in the word frequencies
without any previous assumptions about the dataset. Thus, they
could form the basis of further research focusing on the evaluation
of demographic data estimation from other big data sources, or on
the dynamical processes that result in the patterns found
& [stat.AP]A
Poisson process reparameterisation for Bayesian inference for
Paul Sharkey,
Jonathan A. Tawn
http://arxiv.org/abs/v1&
A common approach to
modelling extreme values is to consider the excesses above a high
threshold as realisations of a non-homogeneous Poisson process.
While this method offers the advantage of modelling using
threshold-invariant extreme value parameters, the dependence
between these parameters makes estimation more difficult. We
present a novel approach for Bayesian estimation of the Poisson
process model parameters by reparameterising in terms of a tuning
parameter m. This paper presents a method for choosing the optimal
value of m that near-orthogonalises the parameters, which is
achieved by minimising the correlation between parameters as
defined by the normalised inverse of the Fisher information
corresponding to the maximum likelihood estimates. This choice of m
ensures more rapid convergence and efficient sampling from the
joint posterior distribution using Markov Chain Monte Carlo
methods. Samples from the parameterisation of interest are then
obtained by a linear transform. Results are presented in the cases
of identically and non-identically distributed models for extreme
rainfall in Cumbria, UK.&
& [stat.AP]An
Overview of Spatial Econometrics
Alexander J.
http://arxiv.org/abs/v1&
This paper offers an
expository overview of the field of spatial econometrics. It first
justifies the necessity of special statistical procedures for the
analysis of spatial data and then proceeds to describe the
fundamentals of these procedures. In particular, this paper covers
three crucial techniques for building models with spatial data.
First, we discuss how to create a spatial weights matrix based on
the distances between each data point in a dataset. Next, we
describe the conventional methods to formally detect spatial
autocorrelation, both global and local. Finally, we outline the
chief components of a spatial autoregressive model, noting the
circumstances under which it would be appropriate to incorporate
each component into a model. This paper seeks to offer a concise
introduction to spatial econometrics that will be accessible to
interested individuals with a background in statistics or
econometrics.&
& [stat.AP]Multi-class Vector
AutoRegressive Models for Multi-store Sales Data
Ines Wilms, Luca
Barbaglia, Christophe Croux
http://arxiv.org/abs/v1&
Retailers use the
Vector AutoRegressive (VAR) model as a standard tool to estimate
the effects of prices, promotions and sales in one product category
on the sales of another product category. Besides, these price,
promotion and sales data are available for not just one store, but
a whole chain of stores. We propose to study cross-category effects
using a multi-class VAR model: we jointly estimate cross-category
effects for several distinct but related VAR models, one for each
store. Our methodology encourages effects to be similar across
stores, while still allowing for small differences between stores
to account for store heterogeneity. Moreover, our estimator is
sparse: unimportant effects are estimated as exactly zero, which
facilitates the interpretation of the results. A simulation study
shows that the proposed multi-class estimator improves estimation
accuracy by borrowing strength across classes. Finally, we provide
three visual tools showing (i) the clustering of stores on
identical cross-category effects, (ii) the networks of product
categories and (iii) the similarity matrices of shared
cross-category effects across stores.&
& [stat.ME]Asymptotic equivalence of
regularization methods in thresholded parameter
Yingying Fan, Jinchi
http://arxiv.org/abs/v1&
High-dimensional data
analysis has motivated a spectrum of regularization methods for
variable selection and sparse modeling, with two popular classes of
convex ones and concave ones. A long debate has been on whether one
class dominates the other, an important question both in theory and
to practitioners. In this paper, we characterize the asymptotic
equivalence of regularization methods, with general penalty
functions, in a thresholded parameter space under the generalized
linear model setting, where the dimensionality can grow up to
exponentially with the sample size. To assess their performance, we
establish the oracle inequalities, as in Bickel, Ritov and Tsybakov
(2009), of the global minimizer for these methods under various
prediction and variable selection losses. These results reveal an
interesting phase transition phenomenon. For polynomially growing
dimensionality, the $L_1$-regularization method of Lasso and
concave methods are asymptotically equivalent, having the same
convergence rates in the oracle inequalities. For exponentially
growing dimensionality, concave methods are asymptotically
equivalent but have faster convergence rates than the Lasso. We
also establish a stronger property of the oracle risk inequalities
of the regularization methods, as well as the sampling properties
of computable solutions. Our new theoretical results are
illustrated and justified by simulation and real data
examples.&
& [stat.ME]Asymptotic properties for
combined $L_1$ and concave regularization
Yingying Fan, Jinchi
http://arxiv.org/abs/v1&
Two important goals
of high-dimensional modeling are prediction and variable selection.
In this article, we consider regularization with combined
$L1$ and concave
penalties, and study the sampling properties of the global optimum
of the suggested method in ultra-high dimensional settings. The
$L1$-penalty provides the minimum regularization needed for
removing noise variables in order to achieve oracle prediction
risk, while concave penalty imposes additional regularization to
control model sparsity. In the linear model setting, we prove that
the global optimum of our method enjoys the same oracle
inequalities as the lasso estimator and admits an explicit bound on
the false sign rate, which can be asymptotically vanishing.
Moreover, we establish oracle risk inequalities for the method and
the sampling properties of computable solutions. Numerical studies
suggest that our method yields more stable estimates than using a
concave penalty alone.&
& [stat.ME]High
dimensional thresholded regression and shrinkage
Zemin Zheng,
Yingying Fan, Jinchi Lv
http://arxiv.org/abs/v1&
High-dimensional
sparse modeling via regularization provides a powerful tool for
analyzing large-scale data sets and obtaining meaningful,
interpretable models. The use of nonconvex penalty functions shows
advantage in selecting important features in high dimensions, but
the global optimality of such methods still demands more
understanding. In this paper, we consider sparse regression with
hard-thresholding penalty, which we show to give rise to
thresholded regression. This approach is motivated by its close
connection with the $L0$-regularization, which can
be unrealistic to implement in practice but of appealing sampling
properties, and its computational advantage. Under some mild
regularity conditions allowing possibly exponentially growing
dimensionality, we establish the oracle inequalities of the
resulting regularized estimator, as the global minimizer, under
various prediction and variable selection losses, as well as the
oracle risk inequalities of the hard-thresholded estimator followed
by a further $L2$-regularization. The risk properties exhibit
interesting shrinkage effects under both estimation and prediction
losses. We identify the optimal choice of the ridge parameter,
which is shown to have simultaneous advantages to both the
$L_2$-loss and prediction loss. These new results and phenomena are
evidenced by simulation and real data
examples.&
& [stat.ME]Innovated scalable efficient
estimation in ultra-large Gaussian graphical
Yingying Fan, Jinchi
http://arxiv.org/abs/v1&
Large-scale precision
matrix estimation is of fundamental importance yet challenging in
many contemporary applications for recovering Gaussian graphical
models. In this paper, we suggest a new approach of innovated
scalable efficient estimation (ISEE) for estimating large precision
matrix. Motivated by the innovated transformation, we convert the
original problem into that of large covariance matrix estimation.
The suggested method combines the strengths of recent advances in
high-dimensional sparse modeling and large covariance matrix
estimation. Compared to existing approaches, our method is scalable
and can deal with much larger precision matrices with simple
tuning. Under mild regularity conditions, we establish that this
procedure can recover the underlying graphical structure with
significant probability and provide efficient estimation of link
strengths. Both computational and theoretical advantages of the
procedure are evidenced through simulation and real data
examples.&
& [stat.ME]Interaction pursuit in
high-dimensional multi-response regression via distance
correlation
Yinfei Kong, Daoji
Li, Yingying Fan, Jinchi Lv
http://arxiv.org/abs/v1&
Feature interactions
can contribute to a large proportion of variation in many
prediction models. In the era of big data, the coexistence of high
dimensionality in both responses and covariates poses unprecedented
challenges in identifying important interactions. In this paper, we
suggest a two-stage interaction identification method, called the
interaction pursuit via distance correlation (IPDC), in the setting
of high-dimensional multi-response interaction models that exploits
feature screening applied to transformed variables with distance
correlation followed by feature selection. Such a procedure is
computationally efficient, generally applicable beyond the heredity
assumption, and effective even when the number of responses
diverges with the sample size. Under mild regularity conditions, we
show that this method enjoys nice theoretical properties including
the sure screening property, support union recovery, and oracle
inequalities in prediction and estimation for both interactions and
main effects. The advantages of our method are supported by several
simulation studies and real data analysis.&
& [stat.ME]Nonparametric hierarchical
Bayesian quantiles
Luke Bornn, Neil
Shephard, Reza Solgi
http://arxiv.org/abs/v1&
Here we develop a
method for performing nonparametric Bayesian inference on
quantiles. Relying on geometric measure theory and employing a
Hausdorff base measure, we are able to specify meaningful priors
for the quantile while treating the distribution of the data
otherwise nonparametrically. We further extend the method to a
hierarchical model for quantiles of subpopulations, linking
subgroups together solely through their quantiles. Our approach is
computationally straightforward, allowing for censored and noisy
data. We demonstrate the proposed methodology on simulated data and
an applied problem from sports statistics, where it is observed to
stabilize and improve inference and
prediction.&
& [stat.ME]The
Distance Precision Matrix: computing networks from nonlinear
relationships
Mahsa Ghanbari,
Julia Lasserre, Martin Vingron
http://arxiv.org/abs/v1&
A fundamental method
of reconstructing networks, e.g. in the context of gene regulation,
relies on the precision matrix (the inverse of the
variance-covariance matrix) as an indicator which variables are
associated with each other. The precision matrix assumes Gaussian
data and its entries are zero for those pairs of variable which are
conditionally independent. Here, we propose the Distance Precision
Matrix which is based on a measure of possibly non-linear
association, the distance covarince. We provide evidence that the
Distance Precision Matrix can successfully compute networks from
non-linear data and does so in a very consistent manner across many
data situations.&
& [stat.ME]The
constrained Dantzig selector with enhanced
consistency
Yinfei Kong, Zemin
Zheng, Jinchi Lv
http://arxiv.org/abs/v1&
The Dantzig selector
has received popularity for many applications such as compressed
sensing and sparse modeling, thanks to its computational efficiency
as a linear programming problem and its nice sampling properties.
Existing results show that it can recover sparse signals mimicking
the accuracy of the ideal procedure, up to a logarithmic factor of
the dimensionality. Such a factor has been shown to hold for many
regularization methods. An important question is whether this
factor can be reduced to a logarithmic factor of the sample size in
ultra-high dimensions under mild regularity conditions. To provide
an affirmative answer, in this paper we suggest the constrained
Dantzig selector, which has more flexible constraints and parameter
space. We prove that the suggested method can achieve convergence
rates within a logarithmic factor of the sample size of the oracle
rates and improved sparsity, under a fairly weak assumption on the
signal strength. Such improvement is significant in ultra-high
dimensions. This method can be implemented efficiently through
sequential linear programming. Numerical studies confirm that the
sample size needed for a certain level of accuracy in these
problems can be much reduced.&
& [stat.ME]Tuning parameter selection in
high dimensional penalized likelihood
Yingying Fan, Cheng
http://arxiv.org/abs/v1&
Determining how to
appropriately select the tuning parameter is essential in penalized
likelihood methods for high-dimensional data analysis. We examine
this problem in the setting of penalized likelihood methods for
generalized linear models, where the dimensionality of covariates p
is allowed to increase exponentially with the sample size n. We
propose to select the tuning parameter by optimizing the
generalized information criterion (GIC) with an appropriate model
complexity penalty. To ensure that we consistently identify the
true model, a range for the model complexity penalty is identified
in GIC. We find that this model complexity penalty should diverge
at the rate of some power of
depending on the tail
probability behavior of the response variables. This reveals that
using the AIC or BIC to select the tuning parameter may not be
adequate for consistently identifying the true model. Based on our
theoretical study, we propose a uniform choice of the model
complexity penalty and show that the proposed approach consistently
identifies the true model among candidate models with asymptotic
probability one. We justify the performance of the proposed
procedure by numerical simulations and a gene expression data
analysis.&
& [stat.ME]What’s the value in more
Timothy Daley,
Andrew D Smith
http://arxiv.org/abs/v1&
The moment-based
non-parametric species richness estimator of Chao is one of the
most widely used estimators for the number of unobserved species in
a sampling experiment. This is due in large part to its simplicity
and robustness. This simplicity can also be a drawback, as it only
uses a small amount of information contained in the observed
experiment, essentially only the first moment. Previous authors,
specifically Harris and Chao, have presented a general moment-based
framework for estimating species richness that includes the Chao
estimator. The application of this framework has been stymied by
both the lack of deep sampling experiments, where higher moments
can be accurately estimated, and the lack of efficient algorithms
to properly use this information. Technological advances have
filled the former void, allowing for sampling experiments orders of
magnitude larger than previously considered. We aim to address the
latter by connecting results from the theory of moment spaces and
Gaussian quadrature to provide a general moment-based
non-parametric estimator of species richness that uses more
information through more moments and is computationally efficient.
We show this estimator performs well and improves upon the Chao
estimator on discrete abundance distributions, the simplest cases
of heterogeneity. We demonstrate the performance on a simulated
populations taken from emerging high-throughput technologies such
as RNA-seq, immune repertoire, and metagenomic
sequencing.&
& [stat.ML]Generalized Sparse Precision
Matrix Selection for Fitting Multivariate Gaussian Random Fields to
Large Data Sets
Sam Davanloo
Tajbakhsh, Necdet Serhat Aybat, Enrique del Castillo
http://arxiv.org/abs/v1&
This paper
generalizes the Sparse Precision matrix Selection (SPS) algorithm,
proposed by Davanloo et al. (2015) for estimating scalar Gaussian
Random Field (GRF) models, to the multivariate, second-order
stationary case under a separable covariance function. Theoretical
convergence rates for the estimated covariance matrix